Mitigating Label Noise on Graph via Topological Sample Selection

TL;DR

This paper introduces a Topological Sample Selection method to improve label noise mitigation in graph neural networks by leveraging topological information, addressing challenges of class boundary nodes and lack of topological measures.

Contribution

The paper proposes a novel TSS method that utilizes graph topology for sample selection, theoretically minimizes expected risk, and outperforms existing baselines.

Findings

TSS effectively reduces label noise impact on GNNs.

Theoretical proof of risk minimization under clean data distribution.

Experimental results show superior performance over state-of-the-art methods.

Abstract

Despite the success of the carefully-annotated benchmarks, the effectiveness of existing graph neural networks (GNNs) can be considerably impaired in practice when the real-world graph data is noisily labeled. Previous explorations in sample selection have been demonstrated as an effective way for robust learning with noisy labels, however, the conventional studies focus on i.i.d data, and when moving to non-iid graph data and GNNs, two notable challenges remain: (1) nodes located near topological class boundaries are very informative for classification but cannot be successfully distinguished by the heuristic sample selection. (2) there is no available measure that considers the graph topological information to promote sample selection in a graph. To address this dilemma, we propose a $\textit{Topological Sample Selection}$ (TSS) method that boosts the informative sample selection process in a graph by utilising topological information. We theoretically prove that our procedure minimizes an upper bound of the expected risk under target clean distribution, and experimentally show the superiority of our method compared with state-of-the-art baselines.